Menstrual and reproductive characteristics of women whose mothers were exposed in utero to diethylstilbestrol (DES)

Transcription

1 Int. J. Epidemiol. Advance Access published May 24, 2006 Published by Oxford University Press on behalf of the International Epidemiological Association Ó The Author 2006; all rights reserved. International Journal of Epidemiology doi: /ije/dyl106 Menstrual and reproductive characteristics of women whose mothers were exposed in utero to diethylstilbestrol (DES) Linda Titus-Ernstoff, 1 * Rebecca Troisi, 1,2 Elizabeth E Hatch, 3 Lauren A Wise, 4 Julie Palmer, 4 Marianne Hyer, 5 Raymond Kaufman, 6 Ervin Adam, 7 William Strohsnitter, 8 Kenneth Noller, 8 Arthur L Herbst, 9 Jennifer Gibson-Chambers, 1 Patricia Hartge 2 and Robert N Hoover 2 Accepted 24 April 2006 Background Methods Results Conclusions In women, prenatal exposure to diethylstilbestrol (DES) is associated with adult reproductive dysfunction. The mouse model, which replicates many DES outcomes, suggests DES causes epigenetic alterations, which are transmissable to daughters of prenatally exposed animals. We report menstrual and reproductive characteristics in a unique cohort comprising daughters of women exposed prenatally to DES. Menstrual and reproductive outcomes and baseline characteristics were assessed by mailed questionnaire in 793 women whose mothers had documented information regarding in utero DES exposure. Mean age at menarche was 12.6 years in both groups, but daughters of the exposed women attained menstrual regularization later (mean age of 16.2 years vs years; P ), and were more likely to report irregular menstrual periods, odds ratio (OR) [95% confidence interval (95% CI )]. A possible association between mothers DES exposure and daughters infertility was compatible with chance, age, and cohort adjusted OR (95% CI ). We found limited evidence that daughters of the exposed had more adverse reproductive outcomes, but daughters of exposed women had fewer live births (1.6) than the (1.9) (P ). The high risk of reproductive dysfunction seen in women exposed to DES in utero was not observed in their daughters, but most women in our cohort have not yet attempted to start their families, and further follow-up is needed to assess their reproductive health. Our findings of menstrual irregularity and possible infertility in third-generation women are preliminary but compatible with speculation regarding transgenerational transmission of DES-related epigenetic alterations in humans. Keywords Diethylstilbestrol, prenatal exposure, maternal exposure, menstruation, reproductive histories, infertility, epigenetic alternations 1 Department of Community and Family Medicine, Dartmouth Medical School, and the Norris Cotton Cancer Center, Lebanon, NH 03756, USA. 2 Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD 20892, USA. 3 Department of Epidemiology and Biostatistics, Boston University School of Public Health, Boston, MA 02118, USA. 4 Slone Epidemiology Center, Boston University School of Public Health, Boston, MA 02215, USA. 5 Information Management Services, Rockville, MD 20852, USA. 6 Department of Obstetrics and Gynecology, Methodist Hospital, Houston, TX 77030, USA. 7 Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, TX 77030, USA. 8 Department of Obstetrics and Gynecology, New England Medical Center, Boston, MA 02111, USA. 9 Department of Obstetrics and Gynecology, University of Chicago, Chicago, IL 60637, USA. * Corresponding author. Dartmouth Medical School and the Norris Cotton Cancer Center, One Medical Center Drive, Lebanon, NH 03756, USA. Linda.Titus-Ernstoff@Dartmouth.edu 1of7

2 2of7 INTERNATIONAL JOURNAL OF EPIDEMIOLOGY Diethylstilbestrol (DES) is a powerful synthetic oestrogen that was administered in high doses to pregnant women under the mistaken belief it would reduce risk of pregnancy complications and losses. Starting in the late 1930s, and continuing through the early 1970s, DES was given to at least two million pregnant women in the US alone. 1 The welldocumented consequences of prenatal DES exposure in adult women include structural anomalies of the reproductive tract, menstrual irregularity, infertility, pregnancy loss, premature delivery, and elevated risk of clear cell cancers of the vagina. 2 8 Studies of reproductive tract tissues in mice indicate that the adverse effects of developmental DES exposure are due to epigenetic mechanisms involving persistent changes in gene expression. 9,10 Epigenetically altered phenotypes may be heritable, 10 and a question of substantial importance is whether the effects of prenatal DES exposure can be transmitted to the next generation of women. Evidence supporting transgenerational transmission is provided by the mouse model, which shows an increased occurrence of reproductive tract tissue anomalies and tumours in the female offspring of mice exposed prenatally to DES In this report, we describe the baseline characteristics, with a particular focus on menstrual and reproductive outcomes, in a unique cohort of daughters born to women who were exposed (or not) to DES in utero. Methods The study described here was approved by the institutional review boards at all participating study centres. Cohort of second-generation men and women In 1992, the National Cancer Institute (NCI) of the United States combined five cohort studies of DES health effects to form the DES follow-up study. Previously followed offspring cohorts were reassembled, and a new cohort was assembled for the first time. The previously followed cohorts included: (i) prenatally exposed offspring of and women who participated between 1951 and 1952 in a clinical trial of DES at the University of Chicago (the Dieckmann cohort); (ii) prenatally exposed women previously enrolled in the National Cooperative Diethylstilbestrol Adenosis Project (DESAD cohort) in the late 1970s and 1980s, their sisters, and age-matched women chosen from the same record sources as the exposed; (iii) prenatally exposed and age-matched men born between 1940 and 1960 at the Mayo Clinic; and (iv) prenatally exposed offspring and their siblings, born between 1952 and 1971 to women attending an infertility clinic in Boston (the Horne cohort). The new cohort consisted of the prenatally exposed and offspring of women who previously participated in the Women s Health Study (WHS), a study of women who took DES during pregnancy. For all cohort members, DES exposure (or lack thereof) was documented by the medical record or physician s note. The characteristics of the original cohorts and subsequent follow-up of the combined offspring cohort have been described previously Cohort of third-generation women In 2000, the NCI initiated a study of adult third-generation women who were born to women enrolled in the five combined cohorts. Potential participants were identified through a review of the parity records of 2323 (1445, 878 ) women who had given birth to at least one daughter. The record review identified 1781 (966 exposed, 815 ) third-generation women who were at least 18 years of age and known to be alive in The five study centres approached the mothers to obtain permission to contact their daughters and to request their daughters current addresses. Mothers gave permission and provided addresses for 898 (50.4%) of the age-eligible third-generation women, including 515 (53.3%) exposed and 383 (47.0%). Enrolment of the third-generation women, which included completing a mailed questionnaire, began in August 2000 and was completed in April Women who did not respond to the initial mailing were sent a second mailing; those who did not respond to the second questionnaire were contacted by telephone, at which time an interview was scheduled or completed if possible. We obtained completed questionnaires from 793 (88%) of 898 third-generation women for whom we had contact information, including 463 (90%) exposed and 330 (86%). The mailed questionnaire queried third-generation women for demographic and personal characteristics, hormone use, menstrual and reproductive histories, and health screening behaviour. Initially, we examined the frequency distributions of variables and means based on the raw data. Analysis of covariance was used to assess the differences in means while adjusting for cohort, or for age and cohort. We used logistic models, adjusted for age (as a continuous factor) and cohort to estimate odds ratios (OR) and 95% confidence intervals (95% CIs) for the association between DES exposure and menstrual or reproductive outcomes. 18 Stratified analyses were used to assess potential interactions. Because some granddaughters had not experienced the outcome of interest, we used a life table approach to calculate the median ages at menstrual regularization (menstrual periods usually predictable within 5 days) and at first pregnancy; log-rank tests were used to assess the difference in the curves for the exposed and. We used proportional hazards models, 19 with age as the time-scale, and adjusted for cohort, to estimate hazards ratios (HRs) for the effect of DES on time to menstrual regularization and first pregnancy. In these analyses, women were censored at the time of questionnaire completion if the event had not occurred. To assess departure from the proportional hazards assumption, we assessed log log survival plots and formally tested interaction terms between DES and age (the study timescale) in relation to the study outcomes. Results The demographic characteristics of the third-generation cohort are shown in Table 1. The daughters of women were younger than daughters of the ; on average, daughters of the exposed were 24.0 years of age, and daughters of the were 25.6 years of age (P, 0.001). Both

3 DES-RELATED OUTCOMES 3of7 Table 1 Demographic characteristics of third-generation women, according to the mother s prenatal DES exposure status Demographic characteristics Age (N 5 463) (N 5 330) < (21.6) 60 (18.2) (35.4) 87 (26.4) (27.2) 88 (26.7) >30 73 (15.8) 95 (28.8) Race/ethnicity White 453 (97.8) 322 (97.6) Other 10 (2.2) 8 (2.4) Education Less than high school graduate 10 (2.2) 15 (4.5) Graduated high school 209 (45.1) 150 (45.5) Post high school 84 (18.1) 59 (17.9) Graduated college or more 158 (34.1) 106 (32.1) Missing 2 (0.4) 0 (0.0) Marital status Single 329 (71.1) 200 (60.6) Married 122 (26.3) 113 (34.2) Divorced/widowed 12 (2.6) 17 (5.2) BMI < (49.7) 154 (46.7) (23.8) 66 (20.0) (9.9) 45 (13.6) (5.6) 20 (6.1) > (10.6) 43 (13.0) Missing 2 (0.4) 2 (0.6) groups were predominately white. Nearly all women had completed high school, and about a third had completed college. The daughters of exposed women were less likely to have been married and tended to have a lower body mass index (BMI) (kg/m 2 ). The proportion of women who had ever used oral contraceptives (OC), and the reasons for using OC were roughly similar for the two groups (Table 2). A slightly higher proportion of exposed women reported current use of OC. On average, both groups started using OC at age 18 years, and duration of use was similar (50.7 months in daughters of the exposed, and 54.3 months in daughters of the, P ). About 10% of women, whether daughters of the exposed or, reported ever using Norplant or Depo Provera (data not shown). Only a small minority of women (3.5% of exposed, 4.8% of ) had ever used non-contraceptive hormones, but the daughters of women used them longer than the daughters of the (16.1 months vs 3.6 months, P ). On average, both groups of women had their first gynaecological examination at age 17 years. The frequency of screening, including gynaecological examinations, colposcopy, Pap tests, mammograms, and clinical breast or general examinations during the previous 5 years was roughly similar for both groups of women (Table 2). If anything, screening was less frequent in daughters of the exposed, compared with the. The mean age at menarche was 12.6 years (median age 13 years) in both groups, and a proportional hazards model indicated that attainment of menarche was similar for the daughters of the exposed and (HR ; 95% CI ). A proportion of women, 9.8% of exposed, and 7.3% of the, had not yet attained regularization of menstrual periods. Among those who had, daughters of the exposed women attained menstrual regularization slightly later, on average, than the ; 16.2 years vs years, respectively (P ). Based on a life table approach, the median ages of menstrual regularization were 15 years in the exposed and 14 years in the (P ). A proportional hazards model suggested that fewer of the exposed had attained regularization at a given age (HR ; 95% CI ). Daughters of the exposed women were more likely to report periods that were usually irregular, age and cohort adjusted OR (95% ), and at least one episode of amenorrhoea (defined here as more than 6 weeks without a menstrual period) during the previous 12 months (OR ; 95% CI ) (Table 3). Further adjustment for BMI did not materially alter the findings with regard to any of the menstrual outcomes. About 5% of women reported ever having an experience of infertility (difficulty getting pregnant for 12 months or more). The age and cohort adjusted OR was 2.19 (95% CI ) for the association between DES and infertility (Table 3). A similar result was obtained when the analyses were confined to those who had ever been married, OR (95% CI ). The age and cohort adjusted OR for seeking medical help for infertility was 1.69 (95% CI ). We were unable, owing to small numbers, to assess specific infertility diagnoses, which were reported by fewer than half of the women who sought related medical care. None of these women attributed infertility solely to the male partner, and only one woman attributed infertility to both the male and the female. We were unable, because of small numbers, to assess the association between DES and infertility in the subgroup whose mothers had self-reported an infertility experience on their own questionnaires. In the subgroup of women whose mothers had not experienced infertility, the association between DES and infertility in the third generation was 1.67 (95% CI ). We explored the possible association with infertility within age strata, adjusted for cohort. Infertility was reported by only 2 women (1 exposed, 1 ) of age <24 years. Among women of age years, there was no evidence that daughters of mothers were more likely to report infertility (OR ; 95% CI ). The association was evident only in the subgroup of women who were of age >30 years at the time they completed their questionnaire; OR (95% CI ). With adjustment for both age and cohort, the OR was 4.21 (95% CI ).

4 4of7 INTERNATIONAL JOURNAL OF EPIDEMIOLOGY Table 2 Hormone use and screening characteristics of third-generation women, according to the mother s prenatal DES exposure status Characteristics Ever use OCs No 99 (21.4) 64 (19.4) Yes 364 (78.6) 265 (80.3) Missing 0 (0.0) 1 (0.3) Reason for OC use Birth control 250 (68.7) 192 (72.5) Regulation of periods 57 (15.7) 37 (14.0) Reduction of menstrual cramps 31 (8.5) 21 (7.9) Other 21 (5.8) 10 (3.8) Missing 5 (1.4) 5 (1.9) Current use of OCs No 158 (43.4) 136 (51.3) Yes 206 (56.6) 129 (48.7) Ever used non-contraceptive hormones No 446 (96.3) 311 (94.2) Yes 16 (3.5) 16 (4.8) Missing 1 (0.2) 3 (0.9) Reason used a Abnormal bleeding/irregular 14 (87.5) 10 (62.6) periods Endometriosis 4 (25.0) 2 (12.5) Other 2 (12.5) 6 (37.5) Number of gynaecological exams in last 5 years None 68 (14.7) 41 (12.4) 1 60 (13.0) 34 (10.3) 2 57 (12.3) 51 (15.5) >3 273 (59.0) 202 (61.2) Missing 5 (1.1) 2 (0.6) Number of colposcopy exams in last 5 years None 353 (76.2) 248 (75.2) 1 47 (10.2) 29 (8.8) 2 20 (4.3) 13 (3.9) >3 20 (4.3) 20 (6.1) Missing 23 (5.0) 20 (6.1) Number of Pap tests in last 5 years None 66 (14.3) 39 (11.8) 1 67 (14.5) 36 (10.9) 2 63 (13.6) 50 (15.2) >3 264 (57.0) 202 (61.2) Missing 3 (0.6) 3 (0.9) Number of mammograms in last 5 years None 434 (93.7) 307 (93.0) 1 17 (3.7) 13 (3.9) 2 4 (0.9) 2 (0.6) >3 1 (0.2) 2 (0.6) Missing 7 (1.5) 6 (1.8) Table 2 continued Characteristics Number of clinical breast exams in last 5 years None 61 (13.2) 37 (11.2) 1 68 (14.7) 34 (10.3) 2 76 (16.4) 55 (16.7) >3 254 (54.9) 199 (60.3) Missing 4 (0.9) 5 (1.5) Number of general exams in last 5 years None 37 (8.0) 19 (5.8) (22.7) 63 (19.1) (24.8) 67 (20.3) >3 200 (43.2) 177 (53.6) Missing 6 (1.3) 4 (1.2) a Respondents reported more than one reason. The majority of women had never been pregnant. After adjustment for age and cohort, there was little association between DES and nulligravidity (OR ; 95% CI ) (Table 3). The average age at first pregnancy among those ever pregnant was similar for daughters of the exposed (21.5 years) and (21.8 years) (P ). Using a life table approach, the median age at first pregnancy was 27 years in the exposed, and 26 years in the, P The proportional hazards model suggested that daughters of the exposed were less likely to experience a first pregnancy at a given age (HR ; 95% CI ). Among women who had ever been pregnant, daughters of the exposed were slightly less likely to report a live birth, but the relationship was compatible with chance (OR ; 95% CI ). On average, and adjusted for age and cohort, the exposed women had fewer live births than the (1.6 vs. 1.9), (P ). The proportion of male offspring was roughly similar for the exposed and women (0.48 vs. 0.54), (P ). The mean birth weight of offspring appeared lower in daughters of the exposed than in the ( g vs g) (P ). The OR for ever having an adverse birth outcome (defined here as a first trimester loss, second trimester loss, ectopic pregnancy, stillbirth, neonatal death, hydatidiform mole) compared with none was compatible with chance (OR ; 95% CI ) (Table 3). Similar proportions of exposed and women reported early (first or second trimester) pregnancy losses, and the proportions were also similar when spontaneous or induced abortions were considered separately (data not shown). The OR was 1.08 (95% CI ) for the relationship between DES and early pregnancy loss (spontaneous or induced), compared with never having had an adverse birth outcome. We found little evidence that daughters of exposed women, compared with daughters of the, were more likely to experience an adverse outcome with their first pregnancy (OR ; 95% CI ). Only a few women (n 5 12) reported a premature birth as a first birth outcome (5.3% of exposed and 2.8% of ).

5 DES-RELATED OUTCOMES 5of7 Table 3 Odds ratios (OR) a and 95% confidence intervals (95% CIs) for the association between the mother s DES exposure status, and menstrual and reproductive characteristics of the daughter Menstrual and reproductive characteristics Irregular periods Discussion n(%) OR (95% CI) No 357 (77.6) 270 (83.9) 1.00 ( ) Yes 103 (22.4) 52 (16.1) 1.54 ( ) Amenorrhoea No 387 (84.3) 284 (87.7) 1.00 ( ) Yes 72 (15.7) 40 (12.3) 1.27 ( ) Infertility experience No 438 (94.6) 313 (94.8) 1.00 ( ) Yes 25 (5.4) 17 (5.2) 2.19 ( ) Sought medical help for infertility No 439 (94.8) 314 (95.2) 1.00 ( ) Yes 24 (5.2) 16 (4.8) 1.69 ( ) Ever been pregnant Yes 150 (32.4) 142 (43.0) 1.00 ( ) No 313 (67.6) 188 (57.0) 1.10 ( ) Ever had a live birth b Yes 103 (69.6) 101 (71.6) 1.00 ( ) No 45 (30.4) 40 (28.4) 0.83 ( ) Ever had an adverse birth outcome b,c No 66 (50.8) 74 (60.7) 1.00 ( ) Any adverse outcome 64 (49.2) 48 (39.3) 1.18 ( ) a Adjusted for age (as a continuous variable) and cohort. b In women who were ever pregnant. c Adverse birth outcomes defined as first trimester loss, second trimester loss, ectopic pregnancy, stillbirth, neonatal death, hydatidiform mole. DES is an established transplacental chemical teratogen and carcinogen in humans. 20 The effects of DES on prenatally exposed women include menstrual irregularity, infertility, adverse reproductive outcomes, anomalies of the reproductive tract, and elevated risk of gynaecological tumours, specifically clear cell adenocarcinoma of the vagina. 1 8 Most of the outcomes seen in women have been replicated in prenatally exposed mice ; thus, the mouse model is useful for investigating DES-related mechanisms. Studies of developmentally exposed rodents indicate that DES exerts its influence on reproductive tissues through epigenetic (non-mutational) mechanisms involving disrupted oestrogen signalling and permanent changes in gene expression, probably due to altered DNA methylation. 20,25,26 Rodent studies have identified altered expression in multiple genes, including those inducible by oestrogen, such as lactoferrin, oestrogen receptor, epidermal growth factor, and specific proto-oncogenes, as well as genes involved with structural development of the reproductive tract, and those governing embryonic development. 9,10,26 36 Of relevance to our study, studies in mice have shown that some DES-related outcomes seen in prenatally exposed animals are also observed in their female descendants Transgenerational effects include reproductive tract tumours, similar to those seen in their prenatally exposed mothers, but may not include increased infertility. 13 To the extent epigenetic heritability is relevant to humans, the mouse model indicates potentially similar effects in the descendants, but allows for some degree of divergence. Studies of the Dieckmann cohort indicate DES does not influence mean age at menarche in the prenatally exposed, 8 or in the daughters of the exposed, based on the mothers reports. 37 Our data, self-reported by the third-generation women, are consistent with the latter finding, indicating a comparable age at menarche for daughters of the exposed and women. However, the Dieckmann studies also indicate that menstrual irregularity and use of hormones to manage menstrual regularity are more common in the prenatally exposed women, compared with the. 37 Similarly, our data indicate that daughters of the exposed women attain menstrual regularity at a slightly later age than daughters of the and are more likely to experience menstrual irregularity. The association between prenatal DES exposure and infertility is well established. 2,3 Our study suggests that infertility may also be more frequent in the daughters of the exposed women, and that DES exposure may exacerbate age-related infertility, a possibility compatible with findings in men who were exposed to DES in utero. 38 None of the women in our study attributed infertility to the male partner, but because most women did not know the reason for their infertility experience, we cannot be certain our findings pertain exclusively to female infertility. Nevertheless, the possible effect observed here could be attributed to male infertility only if the women were more likely than the to marry infertile men, and this seems unlikely. The proportion of third-generation women affected by infertility (5%) in this study was far lower than that observed in the generation of women exposed in utero to DES (30%). 3 In the prenatally exposed women, infertility is primarily due to anatomic anomalies of the uterus or fallopian tubes; other diagnoses, including hormonal/ovulatory problems, play a less striking role. 3 Anatomic and tissue anomalies were not observed in a study of 28 third-generation women, 39 but the number of participants was too small to rule out a low prevalence, and some irregularities (e.g. uterine, tubal) might not be evident on physical examination. Further follow-up is needed to confirm the possible infertility in the thirdgeneration women, and to evaluate specific diagnoses, which may provide insight into DES-related mechanisms. It is well-known that women exposed to DES in utero have increased pregnancy complications and adverse birth outcomes, including ectopic pregnancy, pregnancy loss, and preterm delivery. 4 6 Our data are not conclusive regarding adverse pregnancy outcomes in third-generation women, although daughters of the exposed had somewhat fewer live born children and babies of slightly lower average birth weight. Further follow-up will be essential to assess reproductive outcomes as more of the third-generation women enter their reproductive years. We cannot exclude the possibility that exposed thirdgeneration women with menstrual or reproductive concerns

6 6of7 INTERNATIONAL JOURNAL OF EPIDEMIOLOGY were more likely to participate in our study, or that their mothers were more likely to provide their contact information, which might account for the small differences observed. However, 28% of the daughters of women were unaware of their mothers exposure, and 60% of daughters of the either believed their mothers were or were uncertain of their mothers exposure status. In addition, the medical surveillance was similar for daughters of the exposed and, suggesting the two groups are similar with regard to health consciousness and concerns. We believe our study, based on a unique cohort of women, is the first to show possible transgenerational transmission of phenotypes associated with documented prenatal chemical exposure. Our data provide evidence of menstrual irregularity and delayed menstrual regularization in the daughters of women exposed in utero to DES. We also KEY MESSAGES found possible evidence of increased infertility, particularly in older women. While these findings are preliminary, they are compatible with outcomes seen in the prenatally exposed women and are consistent with speculation that DES exposure during a critical window in human development results in heritable changes in gene function. Although our data provide limited evidence of an increased frequency of adverse pregnancy outcomes in third-generation women, most have not married or attempted to start families, and further follow-up will assess their reproductive experience. Acknowledgement This study was funded by US National Cancer Institute CP and CP Prenatal exposure to DES is known to increase risk of reproductive tract anomalies, reproductive dysfunction, and vaginal adenocarcinoma. The mouse model indicates that the mechanism by which DES exerts its influence includes epigenetic alterations, which may be transmissible to the next generation of females. In a unique cohort of women born to prenatally exposed mothers, we found evidence of menstrual irregularity and infertility, outcomes seen in the women exposed prenatally. Our cohort is young, and most women have not yet attempted to start their families, so further follow-up is needed to assess reproductive health. References 1 Noller KL. In utero exposure to diethylstilbestrol. In: Jones HW III, Wentz AC, Burnett LS (eds). Novak s Textbook of Gynecology. 11th edn. Baltimore: Williams and Wilkins, 1988, pp Senekjian EK, Potkul RK, Frey K, Herbst AL. Infertility among daughters either exposed or not exposed to diethylstilbestrol. Am J Obstet Gynecol 1988;158: Palmer JR, Hatch EE, Rao Sowmya R et al. Infertility among women exposed prenatally to diethylstilbestrol. Am J Epidemiol 2001; 154: Kaufman RH, Adam E, Binder GL, Gerthoffer E. Upper genital tract changes and pregnancy outcome in offspring exposed in utero to diethylstilbestrol. Am J Obstet Gynecol 1980;137: Kaufman RH, Adam E, Hatch EE, Noller KL, Herbst AL, Palmer JR. Continued follow-up of pregnancy outcomes in diethylstilbestrolexposed offspring. Obstet Gynecol 2000;96: Goldberg J, Falcone T. Effect of diethylstilbestrol on reproductive function. Fertil Steril 1999;72: Herbst AL, Scully RE. Adenocarcinoma of the vagina. Association of maternal stilboestrol therapy with tumor appearance in young women. N Engl J Med 1971;284: Bibbo M, Gill WB, Azizi F et al. Follow-up of male and female offspring of mothers. Obstet Gynecol 1977;49: Gray Nelson K, Sakai Y, Eitzman B, Steed T, McLachlan J. Exposure to diethylstilbestrol during a critical development period of the mouse reproductive tract leads to persistent induction of two estrogen-related genes. Cell Growth Differ 1994;5: McLachlan JA. Environmental signaling: what embryos and evolution teach us about endocrine disrupting chemicals. Endocr Res 2001;22: Walker BE. Tumors in female offspring of mice exposed prenatally to diethylstilbestrol. J Natl Cancer Inst 1984;73: Turosov VS, Trukhanova LS, Parfenov YD, Tomatis L. Occurrence of tumors in descendants of CBA male mice prenatally treated with diethylstilbestrol. Int J Cancer 1992;50: Newbold RR, Hanson RB, Jefferson WN, Bullock BC, Hasemann J, McLachlan JA. Increased tumors but uncompromised fertility in the female descendants of mice exposed developmentally to diethylstilbestrol Carcinogenesis 1998;19: Giusti RM, Iwamoto K, Hatch EE. Diethylstilbestrol revisited: a review of the long-term health effects. Ann Intern Med 1995;122: Strohsnitter WC, Noller KL, Hoover RN, Robboy SJ, Palmer JR, Titus-Ernstoff L. Cancer risk in men exposed in utero to diethylstilbestrol. J Natl Cancer Inst 2001;93: Hatch EE, Palmer JR, Titus-Ernstoff L et al. Cancer risk after long term follow-up of women exposed to diethylstilbestrol (DES) in utero. J Am Med Assoc 1998;280: Titus-Ernstoff L, Perez K, Hatch EE et al. Psychosexual characteristics of men and women exposed prenatally to DES (diethylstilbestrol). Epidemiology 2003;14: Breslow NE, Day NE. Statistical methods in cancer research. VII The design and analysis of cohort studies. IARC Scientific Publication No. 82. Lyon: IARC, 1987.

7 DES-RELATED OUTCOMES 7of7 19 Cox DR, Oakes D. Analysis of survival data. London, UK: Chapman Hall, Newbold RR. Lessons learned from perinatal exposure to diethylstilbestrol. Toxicol Appl Pharmacol 2004;199: Walker BE, Uterine tumors in old female mice exposed prenatally to diethylstilbestrol. J Natl Cancer Inst 1983;70: Newbold RR, McLachlan JA. Vaginal adenosis and adenocarcinoma in mice exposed prenatally or neonatally to diethylstilbestrol. Cancer Res 1982;42: McLachlan JA, Newbold RR, Bullock BC. Long-term effects on the female mouse genital tract associated with prenatal exposure to diethylstilbestrol. Cancer Res 1980;40: McLachlan JA, Newbold RR, Shah HC, Hogan M, Doxon RL. Reduced fertility in female mice exposed transplacentally to diethylstilbestrol (DES). Fertil Steril 1982;38: McLachlan JA, Burow M, Chiang T-C, Li SF. Gene imprinting in developmental toxicology: a possible interface between physiology and pathology. Toxicol Lett 2001;120: Li S, Hursting SD, Dzvis BJ, McLachlan JA, Barrett JC. Environmental exposure, DNA methylation, and gene regulation. Lessons from diethylstilbesterol-induced cancers. Ann N Y Acad Sci 2003;983: Li JJ, Hansman R, Newbold R, Davis B, McLachlan JA, Barrett JC. Developmental exposure to diethylstilbestrol elicits demethylation of estrogen-responsive lactoferrin gene in mouse uterus. Cancer Res 1997;57: Couse JF, Dixon D, Yates M et al. Estrogen receptor-a knockout mice exhibit resistance to the developmental effects of neonatal diethylstilbestrol exposure on the female reproductive tract. Dev Biol 2001;238: Mericskay M, Carta L, Sassoon D. Diethylstilbestrol exposure in utero: A paradigm for mechanisms leading to adult disease. Birth Defects Res A Clin Mol Teratol 2005;73: Kamiya K, Sato T, Nishimura N, Goto Y, Kano K, Iguchi T et al. Expression of estrogen receptor and proto-oncogene messenger ribonucleic aids in reproductive tracts of neonatally diethylstilbestrolexposed female mice with or without post-pubertal estrogen administration. Exp Clin Endocrinol Diabetes 1996;104: Yamashita S, Takayanagi A, Shimizu N. Effects of neonatal diethylstilbestrol exposure on c-fos and c-jun protooncogene expression in the mouse uterus. Histol Histopathol 2001;16: Falck L, Forsberg J-G. Immunohistochemical studies on the expression and estrogen dependency of EGF and its receptor and C-fos proto-oncogene in the uterus and vagina of normal and neonatally estrogen-treated mice. Anat Rec 1996;245: Ma L, Benson GV, Lim H, Dey SK, Maas RL. Abdominal B (AbdB) Hoxa genes: regulation in adult uterus by estrogen and progesterone and repression in mullerian duct by the synthetic estrogen diethylstilbestrol (DES). Dev Biol 1998;197: Block K, Kardana A, Igarashi P, Taylor HS. In utero diethylstilbestrol (DES) exposure alters Hox gene expression in the developing mullerian system. FASEB J 2000;4: Sato T, Ohta Y, Okamura H, Hayashi S, Iguchi T. Estrogen receptor (ER) and its messenger ribonucleic acid expression in the genital tract of female mice exposed neonatally to tamoxifen and diethylstilbestrol. Anat Rec 1996;44: Okada A, Sato T, Ohta Y, Buchanan DL, Iguchi T. Effect of diethylstilbestrol on cell proliferation and expression of epidermal growth factor in the developing female rat reproductive tract. J Endocrinol 2001;170: Wilcox AJ, Umbach DM, Hornsby PP, Herbst AL. Age at menarche among diethylstilbestrol granddaughters. Am J Obstet Gynecol 1995;73: Perez K, Titus-Ernstoff L, Hatch EE et al. Reproductive outcomes in men with prenatal exposure to DES. Fertil Steril 2005;84: Kaufman RH, Adam E. Findings in female offspring of women exposed in-utero to diethylstilbestrol. Obstet Gynecol 2002;99: